Latest Post

Ballots Don’t Reflect the People’s Will

(This previously appeared in Issue 2 of the Vermont Movement News in 2015)

The recent election of Governor Shumlin has caused many commentators to call for a change in how our statewide officials are elected, but few have challenged the many assumptions built into our system of elections. For example, one lawmaker claimed that “[a] core pillar of democracy is that the person who gets the most votes, wins.” While some might have an attachment to the specific mechanism that we currently use to elect our representatives in government, this mechanism is most certainly not a core pillar of democracy. The only time that our vote-for-one-person and whoever-gets-the-most-votes system works reasonably well is when we have exactly two choices and we must select one. Even then, voters dissatisfied with the choices often lack the ability to express “None of the Above”!

The assumption that we must fill in exactly one bubble and that the candidate with the most bubbles filled in next to their name is insidious. Political parties have been built solely to allow people to use their ballots as a medium of protest, and independent candidates often run for the same purpose. With these candidates on the ballot, a voter can vote for someone who will almost certainly not win to indicate that the voter is dissatisfied with the electoral process and hopefully, with a critical mass of likeminded voters, communicate some message to some person somewhere. This sort of protest—which I admit I’ve engaged in—is attractive in part because the current voting system does not allow voters to express their actual preferences, which may be a factor in the miserable voter turnout in Vermont and throughout the United States.

Political campaigns are executed pragmatically and representative district lines are redrawn to take advantage of the idiosyncrasies of the voting system. This means that candidates often pander to certain constituent groups while completely ignoring others. With declining voter turnout, a candidate really only needs to make sure that more of their supporters show up than their opponent’s, and that means that little attention is paid to the remaining independent, disenfranchised, or simply busy voters. As we saw in the 2014 election, this can result in a tiny minority of registered voters (not to mention an even tinier minority of all people eligible to vote) electing someone who is supposed to represent an entire district.

Much ink—academic as well as journalistic—has been spilled about which alternative election system would be a better replacement for what we have now. I’m not a cheerleader for any particular system, as they all have their strengths and weaknesses. What they all have in common is that every one of them is profoundly better than what we use now. We have remained comfortable with a system that some of the time results in the election of the worst possible candidate, and we stick with it presumably because that’s the way we’ve done elections for a long time.

First notice that voting is made up of two distinct parts: the method used by voters to communicate their preferences (academics call this the balloting method) and the method used to determine the winner based on the voters’ preferences (called the decision rule). Ideally, voters would be able to communicate a detailed explanation of who they thought would be best for the job and why, but people don’t generally want to do that and compiling such preferences would be nightmarishly complex and subjective.

The balloting method we currently use in most elections is vote-for-one, and in multi-member districts like many State Senate and State Representative districts, it’s vote-for-the-number-of-winners. Other balloting methods that are more expressive of voter preferences include approval voting, range voting, and candidate ranking. Approval voting allows for voters to fill in the bubble next to every candidate who they think would do a good job, not just the number of possible winners. Range voting offers voters the ability to give a numeric score to each candidate (say, 0 through 5), so that if they really like the first two candidates, somewhat like the third, and really dislike the fourth, they could vote 5, 5, 3, and 0, respectively. Candidate ranking allows the voter to place all candidates in a strict order, so that four candidates A, B, C, and D might (for example) be ranked C, A, D, B if the voter preferred C to A, A to D, and D to B.

How the winner is determined based in large part upon how much information the ballots provide. We are most familiar with the plurality decision rule: the candidate that receives the most votes wins. When applied to an approval or range ballot, the winner is calculated in exactly the same way – the person who was “approved” by the most voters or who received the highest total score wins. Calculating the winner of a ranked ballot with the Instant Runoff Voting (IRV) decision rule starts by considering if any single candidate was put in first place by more than half of the voters. Without diving into the details too much, voters’ second and third choice candidates are incrementally considered until a single candidate emerges with a majority, even though this may not be the candidate who initially received the most first-place votes.

How many people would have thought that Peter Diamondstone, Scott Milne, and Peter Shumlin were equally inept or equally qualified for the office of Governor? We can never know with the current vote-for-one highest-vote-total-takes all system. Even if we try to be savvy and look for patterns in the ballots themselves, it’s hard to draw strong conclusions, though some of my preliminary findings suggest that pro-single-payer voters did not leave Shumlin in droves; very few voters cast party-line ballots; and some voters may be mixing up the Liberty Union and Libertarian parties. In the case of deciding who was to be the Governor, legislators were left crafting their own reasons or excuses for voting the way that they did (which fell almost exactly on party lines) without the benefit of doing any sort of statistical analysis of the ballots, using only the little bit of extra information that they could glean from town-by-town and district-by-district vote totals. This is no way to reflect the opinions of the people.

Aside from the increasing influence of corporate money in our elections, our inability to elect representatives that truly represent us is due in large part to our decision to continue using a flawed voting system. Let’s give ourselves the tools to make our choices clear and unambiguous.


Get every new post delivered to your Inbox.